Search Results (230)

Spike structures appear in several phenomena, whereas spike trains (STs) are of particular importance, since they can carry temporal encoding of information. Regarding the STs of the biological neuron type, several models have already been proposed. While existing models effectively simulate spike generation, they fail to capture the dynamics of high-frequency spontaneous membrane potential fluctuations observed during relaxation intervals between consecutive spikes, dismissing them as random noise. This is eventually an important drawback because it has been shown that, in real data, these spontaneous fluctuations are not random noise. In this work, we suggest an ST production mechanism based on the appropriate coupling of two specific intermittent maps, which are nonlinear first-order difference equations. One of these maps presents small variation in low amplitude values and, at some point, bursts to high values, whereas the other presents the inverse behavior, i.e., from small variation in high values, bursts to low values. The suggested mechanism proves to be able to generate the above-mentioned spontaneous membrane fluctuations possessing the associated dynamical properties observed in real data. Moreover, it is shown to produce spikes that present spike threshold, sharp peak and the hyperpolarization phenomenon, which are key morphological characteristics of biological spikes. Furthermore, the inter-spike interval distribution is shown to be a power law, in agreement with published results for ST data produced by real biological neurons. The use of the suggested mechanism for the production of other types of STs, as well as possible applications, are discussed. Full article

The explosive growth of power load data has led to a substantial presence of abnormal data, which significantly reduce the accuracy of power system operation planning, load forecasting, and energy usage analysis. To address this issue, a novel improved GAN–Transformer model is proposed, leveraging the adversarial structure of the generator and discriminator in Generative Adversarial Networks (GANs). To provide the model with a suitable feature dataset, One-hot encoding is introduced to label different categories of abnormal power load data, enabling staged mapping and training of the model with the labeled dataset. Experimental results demonstrate that the proposed model accurately identifies and classifies mutation anomalies, sustained extreme anomalies, spike anomalies, and transient extreme anomalies. Furthermore, it outperforms traditional methods such as LSTM-NDT, Transformer, OmniAnomaly and MAD-GAN in Overall Accuracy, Average Accuracy, and Kappa coefficient, thereby validating the effectiveness and superiority of the proposed anomaly detection and classification method. Full article

Background: With the development of artificial intelligence, memristors have become an ideal choice to optimize new neural network architectures and improve computing efficiency and energy efficiency due to their combination of storage and computing power. In this context, spiking neural networks show the ability to resist Gaussian noise, spike interference, and AC electric field interference by adjusting synaptic plasticity. The anti-interference ability to spike neural networks has become an important direction of electromagnetic protection bionics research. Methods: Therefore, this research constructs two types of spiking neural network models with LIF model as nodes: VGG-SNN and FCNN-SNN, and combines pruning algorithm to simulate network connection damage during the training process. By comparing and analyzing the millimeter wave radar human motion dataset and MNIST dataset with traditional artificial neural networks, the anti-interference performance of spiking neural networks and traditional artificial neural networks under the same probability of edge loss was deeply explored. Results: The experimental results show that on the millimeter wave radar human motion dataset, the accuracy of the spiking neural network decreased by 5.83% at a sparsity of 30%, while the accuracy of the artificial neural network decreased by 18.71%. On the MNIST dataset, the accuracy of the spiking neural network decreased by 3.91% at a sparsity of 30%, while the artificial neural network decreased by 10.13%. Conclusions: Therefore, under the same network connection damage conditions, spiking neural networks exhibit unique anti-interference performance advantages. The performance of spiking neural networks in information processing and pattern recognition is relatively more stable and outstanding. Further analysis reveals that factors such as network structure, encoding method, and learning algorithm have a significant impact on the anti-interference performance of both. Full article

Background/Objectives: Repeated spiking and serving movements in volleyball can lead to alterations in shoulder range of motion among athletes, potentially increasing the risk of shoulder instability and injury. Hence, assessing and understanding the shoulder range of motion of volleyball players is a critical concern. Therefore, this study aimed to understand and evaluate the bilateral shoulder joint range of motion (ROM) in high-school male volleyball athletes and to discover the adaptation characteristics. Methods: Forty high-school male volleyball athletes participated in this study. Shoulder ROM measurements were taken via video with an iPhone 12 Pro Max, and we analyzed the ROM data using Kinovea software (Version 0.9.5) for both the dominant and non-dominant side. The shoulder ROM measurements included shoulder hyper-extension (SE), flexion (SF), internal rotation (IR), external rotation (ER), horizontal adduction (Sadd), and horizontal abduction (Sabd). After taking shoulder ROM measurements, the total rotational range of motion (TROM) was calculated based on the participants’ shoulder internal rotation and external rotation data, and we calculated the incidence of glenohumeral internal rotation deficiency (GIRD) among participants. Paired samples t-tests were used to analyze shoulder ROM differences between the dominant and non-dominant side. Results: The dominant side of the shoulder showed significantly lower internal rotation (dominant side: 42.17 ± 11.23°; non-dominant side: 52.14 ± 10.46°; p = 0.000) and total rotational ROM (dominant side: 137.11 ± 13.09°; non-dominant side: 141.96 ± 13.22°; p = 0.021) compared to the non-dominant side. Conversely, the dominant side of the shoulder exhibited significantly greater external rotation (dominant side: 94.96 ± 10.02°; non-dominant side: 89.83 ± 7.84°; p = 0.001) and shoulder horizontal adduction (dominant side: 44.87 ± 8.10°; non-dominant side: 39.60 ± 7.24°; p = 0.000) than the non-dominant side. No significant differences were found in other measured parameters. The incidence of glenohumeral internal rotation deficiency (GIRD) among all subjects was 37.5%. Conclusions: High-school male volleyball athletes in this study exhibited tightness in the posterior shoulder of their dominant side, indicating specific adaptations in shoulder ROM and a considerable prevalence of GIRD, observed in approximately one-quarter of the athletes. In conclusion, these data suggest that stretching and eccentric muscle training focusing on the posterior shoulder have potential value in mitigating these adaptations and reducing the risk of shoulder injuries. Full article

This study compares three machine learning and deep learning models—Support Vector Regression (SVR), Recurrent Neural Networks (RNN), and Long Short-Term Memory (LSTM)—for predicting market prices using the NGX All-Share Index dataset. The models were evaluated using multiple error metrics, including Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Square Error (RMSE), Mean Percentage Error (MPE), and R-squared. RNN and LSTM were tested with both 30 and 60-day windows, with performance compared to SVR. LSTM delivered better R-squared values, with a 60-day LSTM achieving the best accuracy (R-squared = 0.993) when using a combination of endogenous market data and technical indicators. SVR showed reliable results in certain scenarios but struggled in fold 2 with a sudden spike that shows a high probability of not capturing the entire underlying NGX pattern in the dataset correctly, as witnessed by the high validation loss during the period. Additionally, RNN faced the vanishing gradient problem that limits its long-term performance. Despite challenges, LSTM’s ability to handle temporal dependencies, especially with the inclusion of On-Balance Volume, led to significant improvements in prediction accuracy. The use of the Optuna optimisation framework further enhanced model training and hyperparameter tuning, contributing to the performance of the LSTM model. Full article

Plyometric training is known to improve jump height in volleyball, but few studies address athletes’ perception and awareness of its benefits. This gap limits its full potential for enhancing performance. This study examines young non-elite volleyball athletes’ awareness of plyometric training effects. A sample of 24 athletes (mean age 18.3 ± 3.8 years) was divided into an experimental group (EXP) and a control group (CON), each with 12 participants. The EXP group underwent plyometric training, while the CON group performed basic technical exercises. Performance and perceptions were assessed using the Spike Jump Test and surveys at pre-, mid-, and post-training phases. The EXP group showed significant vertical jump improvement, from a pre-training mean of 30.14 cm to 32.22 cm post-training, confirmed by the Friedman test (p = 0.00). In contrast, the CON group showed no significant changes (p = 0.47). Perception scores in the EXP group improved significantly, from 3.33 to 4.16, indicating enhanced awareness of plyometric training benefits, whereas the CON group showed no significant changes (p = 0.35). These findings highlight the dual benefits of plyometric training in improving both jump performance and awareness of its effectiveness, emphasizing the value of integrating perception into training for volleyball athletes. Full article

The cart-pole application is a well-known control application that is often used to illustrate reinforcement learning algorithms with conventional neural networks. An implementation of the application from OpenAI Gym is ubiquitous and popular. Spiking neural networks are the basis of brain-based, or neuromorphic computing. They are attractive, especially as agents for control applications, because of their very low size, weight and power requirements. We are motivated to help researchers in neuromorphic computing to be able to compare their work with common benchmarks, and in this paper we explore using the cart-pole application as a benchmark for spiking neural networks. We propose four parameter settings that scale the application in difficulty, in particular beyond the default parameter settings which do not pose a difficult test for AI agents. We propose achievement levels for AI agents that are trained with these settings. Next, we perform an experiment that employs the benchmark and its difficulty levels to evaluate the effectiveness of eight neuroprocessor settings on success with the application. Finally, we perform a detailed examination of eight example networks from this experiment, that achieve our goals on the difficulty levels, and comment on features that enable them to be successful. Our goal is to help researchers in neuromorphic computing to utilize the cart-pole application as an effective benchmark. Full article

Background/Objectives: One of the problematic situations dermatologists face with their patients is communicating dismal diagnoses. Examples are the diagnosis and prognosis of skin cancers like melanoma and Merkel cell carcinoma and the disclosure of the chronic nature of a disease that requires long-term therapies or can lead to scarring or disfiguring conditions. Likewise, receiving a diagnosis of a sexually transmitted infection can be a shocking event that can also put into question the patient’s relationship with his/her partner/partners. Some oncology and internal medicine protocols have been developed to support delivering distressing information. Regrettably, no consensus guidelines exist in dermatology, sexually transmitted infections, or other medical specialties. Methods: The protocols available in the literature to guide the disclosure of a dismal diagnosis have been reviewed in the present work. Results: The different protocols consist of several steps, from 5 to 13, and most of them are summarized by acronyms, such as “SPIKES”, “ABCDE”, and “BREAKS”. The frameworks are listened to and explained in the manuscript. Conclusions: These communication models are suggested to be adapted to dermatology and sexually transmitted infections. Indeed, several studies demonstrated that training in communication skills and techniques to facilitate breaking bad news may improve patient satisfaction and physician comfort. Full article

Epilepsy is a group of neurological disorders characterized by epileptic seizures, and it affects tens of millions of people worldwide. Currently, the most effective diagnostic method employs the monitoring of brain activity through electroencephalogram (EEG). However, it is critical to predict epileptic seizures in patients prior to their onset, allowing for the administration of preventive medications before the seizure occurs. As a pivotal application of artificial intelligence in medical treatment, learning the features of EEGs for epilepsy prediction and detection remains a challenging problem, primarily due to the presence of intra-class and inter-class variations in EEG signals. In this study, we propose the spatio-temporal EEGNet, which integrates contractive slab and spike convolutional deep belief network (CssCDBN) with a self-attention architecture, augmented by dual-task learning to address this issue. Initially, our model was designed to extract high-order and deep representations from EEG spectrum images, enabling the simultaneous capture of spatial and temporal information. Furthermore, EEG-based verification aids in reducing intra-class variation by considering the time correlation of the EEG during the fine-tuning stage, resulting in easier inference and training. The results demonstrate the notable efficacy of our proposed method. Our method achieved a sensitivity of 98.5%, a false-positive rate (FPR) of 0.041, a prediction time of 50.92 min during the epilepsy prediction task, and an accuracy of 94.1% during the epilepsy detection task, demonstrating significant improvements over current state-of-the-art methods. Full article

Background: Migrant construction workers involved in building infrastructure for mega-sporting events face elevated risks of illness and death. However, specific health outcomes for these workers have not been systematically reviewed, limiting opportunities to identify and address their challenges. Methods: This study systematically reviewed health outcomes among migrant construction workers involved in mega-sporting events. Results: 89 eligible studies involving 23,307 workers were identified. Of these, only 11 directly addressed specific health outcomes, including heat stress, occupational fatalities, and sexually transmitted infections. Notably, increased heat exposure during peak construction phases and the proximity of deadlines for mega-sporting events were correlated with elevated rates of occupational fatalities. Other key adverse factors impacting migrant construction workers’ health included an observed correlation between the timing of mega-sporting events and increased occupational fatalities, the involvement of labor recruiters, and shifting health and safety responsibilities among stakeholders (e.g., host states, event organizers, contractors, and recruitment agencies). Positive outcomes were observed when workers voluntarily engaged in non-mandatory safety activities, such as safety training programs and awareness meetings. Conclusions: There is a critical need for longitudinal and comparative studies to comprehensively examine the health of migrant workers throughout all stages of their journey, from pre-migration to return. This review underscores the urgency of prioritizing evidence-based policies that address unique health risks in this population, including mitigation of heat stress and enforcement of occupational safety standards, particularly amid construction spikes preceding mega-sporting events. Recommendations: Future research should prioritize understanding the unique health challenges faced by migrant workers to inform policy making, develop effective interventions, and implement best practices to improve their health and well-being. Full article

The rapid growth of Internet of Things (IoT) devices has significantly increased reliance on sensor-generated data, which are essential to a wide range of systems and services. Wireless sensor networks (WSNs), crucial to this ecosystem, are often deployed in diverse and challenging environments, making them susceptible to faults such as software bugs, communication breakdowns, and hardware malfunctions. These issues can compromise data accuracy, stability, and reliability, ultimately jeopardizing system security. While advanced sensor fault detection methods in WSNs leverage a machine learning approach to achieve high accuracy, they typically rely on centralized learning, and face scalability and privacy challenges, especially when transferring large volumes of data. In our experimental setup, we employ a decentralized approach using federated learning with long short-term memory (FedLSTM) for sensor fault detection in WSNs, thereby preserving client privacy. This study utilizes temperature data enhanced with synthetic sensor data to simulate various common sensor faults: bias, drift, spike, erratic, stuck, and data-loss. We evaluate the performance of FedLSTM against the centralized approach based on accuracy, precision, sensitivity, and F1-score. Additionally, we analyze the impacts of varying the client participation rates and the number of local training epochs. In federated learning environments, comparative analysis with established models like the one-dimensional convolutional neural network and multilayer perceptron demonstrate the promising results of FedLSTM in maintaining client privacy while reducing communication overheads and the server load. Full article

This study investigates neural activity changes in the medial prefrontal cortex (mPFC) of a lipopolysaccharide (LPS)-induced acute depression mouse model using flexible polymer multichannel electrodes, local field potential (LFP) analysis, and a convolutional neural network-long short-term memory (CNN-LSTM) classification model. LPS treatment effectively induced depressive-like behaviors, including increased immobility in the tail suspension and forced swim tests, as well as reduced sucrose preference. These behavioral outcomes validate the LPS-induced depressive phenotype, providing a foundation for neurophysiological analysis. Flexible polymer-based electrodes enabled the long-term recording of high-quality LFP and spike signals from the mPFC. Time-frequency and power spectral density (PSD) analyses revealed a significant increase in theta band (3–8 Hz) amplitude under depressive conditions. Using theta waveform features extracted via empirical mode decomposition (EMD), we classified depressive states with a CNN-LSTM model, achieving high accuracy in both training and validation sets. This study presents a novel approach for depression state recognition using flexible polymer electrodes, EMD, and CNN-LSTM modeling, suggesting that heightened theta oscillations in the mPFC may serve as a neural marker for depression. Future studies may explore theta coupling across brain regions to further elucidate neural network disruptions associated with depression. Full article

Objective: This study evaluated the effects of advanced footwear technology (AFT) spikes on running performance measures, spatiotemporal variables, and perceptive parameters on different surfaces (track and grass). Methods: Twenty-seven male trained runners were recruited for this study. In Experiment 1, participants performed 12 × 200 m at a self-perceived 3000 m running pace with a recovery of 5 min. Performance (time in each repetition), spatiotemporal, and perceptive parameters were measured. In Experiment 2, participants performed 8 × 5 min at 4.44 m/s while energy cost of running (W/kg), spatiotemporal, and perceptive parameters were measured. In both experiments the surface was randomized and mirror order between spike conditions (Polyether Block Amide (PEBA) and PEBA + Plate) was used. Results: Experiment 1: Runners were faster on the track (p = 0.002) and using PEBA + Plate spike (p = 0.049). Experiment 2: Running on grass increased energy cost (p = 0.03) and heart rate (p < 0.001) regardless of the spike used, while PEBA + Plate spike reduced respiratory exchange ratio (RER) (p = 0.041). Step frequency was different across surfaces (p < 0.001) and spikes (p = 0.002), with increased performance and comfort perceived with PEBA + Plate spikes (p < 0.001; p = 0.049). Conclusions: Running on the track surface with PEBA + Plate spikes enhanced auto-perceived 3000 m running performance, showed lower RER, and improved auto-perceptive comfort and performance. Running on grass surfaces increased energy cost and heart rate without differences between spike conditions. Full article

This randomized controlled trial investigated the effects of neuro-athletic training (NAT) on flexibility, spike speed, and upper extremity stability in elite volleyball players. Thirty professional male athletes aged 18–23 years old (mean age of 19.5 ± 1.77 years old in the NAT group and 19.8 ± 1.87 years old in the control group) participated, with 26 completing this study. The participants were randomly assigned into an NAT intervention group or a control group continuing traditional training. Both groups trained three days per week for eight weeks, with the NAT program targeting neuromuscular adaptations while maintaining equal total training durations. Flexibility was assessed using the Sit and Reach Test, spike speed was evaluated using the Pocket Radar Ball Coach, and upper extremity stability was measured using the Closed Kinetic Chain Upper Extremity Stability Test (CKCUEST). The NAT group demonstrated significant improvements across all performance metrics. Flexibility increased significantly (p = 0.040; Cohen’s d = 0.845), indicating improved range of motion and musculoskeletal adaptability. Spike speed showed a highly significant improvement (p < 0.001; Cohen’s d = 1.503), reflecting enhanced neuromuscular coordination and power. Similarly, upper extremity stability exhibited substantial gains (p = 0.002; Cohen’s d = 1.152), highlighting improved shoulder stability and motor control. In contrast, the control group did not show statistically significant changes in their flexibility (p = 0.236; Cohen’s d = 0.045), spike speed (p = 0.197; Cohen’s d = 0.682), or upper extremity stability (p = 0.193; Cohen’s d = 0.184). Between-group comparisons confirmed the superiority of the NAT intervention, with significant differences across all metrics (p-values ranging from 0.040 to <0.001) and effect sizes spanning from moderate to large (Cohen’s d = 0.845–1.503). These findings demonstrate the effectiveness of NAT in enhancing volleyball-specific performance metrics, emphasizing its potential to target neuromuscular adaptations for improved flexibility, power, and stability. Future studies should explore the long-term effects of NAT and its applicability across various sports disciplines. Full article

This paper presents a novel approach to in situ memristive learning by training spiking neural networks (SNNs) entirely within the circuit using memristor emulators in SPICE. The circuit models neurons using Lapicque neurons and employs pulse-based spike encoding to simulate spike-timing-dependent plasticity (STDP), a key learning mechanism in SNNs. The Lapicque neuron model operates according to the Leaky Integrate-and-Fire (LIF) model, which is used in this study to model spiking behavior in memristor-based SNNs. More exactly, the first memristor emulator in PySpice, a Python library for circuit simulation, was developed and integrated into a memristive circuit capable of in situ learning, named the “In Situ Memristive Learning Method for Pattern Classification”. This novel technique enables time-based computation, where neurons accumulate incoming spikes and fire once a threshold is reached, mimicking biological neuron behavior. The proposed method was rigorously tested on three diverse datasets: XPUE, a custom non-dominating 3 × 3 image dataset; a 3 × 5 digit dataset ranging from 0 to 5; and a resized 10 × 10 version of the Modified National Institute of Standards and Technology (MNIST) dataset. The neuromorphic circuit achieved successful pattern learning across all three datasets, outperforming comparable results from other in situ training simulations on SPICE. The learning process harnesses the cumulative effect of memristors, enabling the network to learn a representative pattern for each label efficiently. This advancement opens new avenues for neuromorphic computing and paves the way for developing autonomous, adaptable pattern classification neuromorphic circuits. Full article